Diagnostic value of machine learning-based computed tomography texture analysis for differentiating multiple myeloma from osteolytic metastatic bone lesions in the peripheral skeleton.

OBJECTIVES: To report the diagnostic performance of machine learning-based CT texture analysis for differentiating multiple myeloma from osteolytic metastatic bone lesions in the peripheral skeleton. METHODS: We retrospectively evaluated 172 patients with multiple myeloma (n = 70) and osteolytic metastatic bone lesions (n = 102) in the peripheral skeleton. Two radiologists individually used two-dimensional manual segmentation to extract texture features from non-contrast CT. In total, 762 radiomic features were extracted. Dimension reduction was performed in three stages: inter-observer agreement analysis, collinearity analysis, and feature selection. Data were randomly divided into training (n = 120) and test (n = 52) groups. Eight machine learning algorithms were used for model development. The primary performance metrics were the area under the receiver operating characteristic curve and accuracy. RESULTS: In total, 476 of the 762 texture features demonstrated excellent interobserver agreement. The number of features was reduced to 22 after excluding those with strong collinearity. Of these features, six were included in the machine learning algorithms using the wrapper-based classifier-specific technique. When all eight machine learning algorithms were considered for differentiating multiple myeloma from osteolytic metastatic bone lesions in the peripheral skeleton, the area under the receiver operating characteristic curve and accuracy were 0.776-0.932 and 78.8-92.3%, respectively. The k-nearest neighbors model performed the best, with the area under the receiver operating characteristic curve and accuracy values of 0.902 and 92.3%, respectively. CONCLUSION: Machine learning-based CT texture analysis is a promising method for discriminating multiple myeloma from osteolytic metastatic bone lesions.

The accuracy of three-dimensional ultrasonography in the diagnosis of Müllerian duct anomalies and its concordance with magnetic resonance imaging.

This retrospective study was performed to comparatively evaluate the diagnostic accuracies of three-dimensional ultrasonography (3D-US) and magnetic resonance imaging (MRI) for identification of Müllerian duct anomalies (MDAs). A total of 27 women with suspected MDAs underwent gynaecological examination, 2D-US, 3D-US and MRI, respectively. The MDAs were classified with respect to the European Society of Human Reproduction and Embryology-European Society for Gynaecological Endoscopy (ESHRE/ESGE) and American Society of Reproductive Medicine (ASRM) systems. Based on the ESHRE/ESGE classification, there was a discrepancy for only one patient between US and MRI. Thus, the concordance between US and MRI was 26/27 (96.3%). With respect to ASRM classification, there was a disagreement between MRI and 3D-US in three patients, thus the concordance between MRI and 3D-US was 24/27 (88.9%). To conclude, the 3D-US has a good level of agreement with MRI for recognition of MDAs.Impact StatementWhat is already known on this subject? Müllerian duct anomalies (MDAs) are relatively common malformations of the female genital tract and they may adversely affect the reproductive potential. The establishment of accurate and timely diagnosis of these malformations is critical to overcome clinical consequences of MDAs.What the results of this study add? The concordance between US and MRI for diagnosis of MDAs based on ESHRE-ESGE classification and ASRM were 96.3% and 88.9%, respectively. These results indicate that 3D US has a satisfactory level of diagnostic accuracy for MDAs and it can be used in conjunction with MRI. Minimisation of diagnostic errors is important to improve reproductive outcome and to avoid unnecessary surgical interventions.What the implications are of these findings for clinical practice and/or further research? Efforts must be spent to eliminate the discrepancies between the clinical and radiological diagnosis of MDAs. Further trials should be implemented for establishment and standardisation of radiological images for identification and classification of MDAs.

Evaluation of computed tomography images of calcaneus for estimation of sex.

Estimation of sex from bone remnants after medicolegal issues or mass disasters constitute a challenge. Our objective was to investigate if morphometric measurements performed on computed tomography (CT) images of the calcaneus can be useful for sex estimation. The measurements were performed on CT images of calcaneus bones of 489 patients (203 women, 286 men). The parameters under investigation were maximum length (MAXL), load arm length (LAL), minimum breadth (MINB), body height (BH), maximum height (MAXH), dorsal articular facet length (DAFL), dorsal articular facet breadth (DAFB), and cuboidal facet height (CFH). All parameters were remarkably increased in men compared to women (p < 0.001, for all). The highest performances for sex estimation were displayed by DAFL (92.8%) and BH (92.6%). The concomitant use of variables MAXL, BH, CFH, and DAFL revealed accuracy as high as 96% for gender discrimination. We suggest that morphometric measurements performed on CT images of the calcaneus can provide useful data for the estimation of sex.

Evaluation of the Ossification Stage of Proximal Humeral Epiphysis by 1.5-T Magnetic Resonance Imaging for Determination of Skeletal Age.

PURPOSE: The aim of the study was to investigate whether the skeletal age can be evaluated assessed by means of analysis of ossification stage at proximal humeral epiphysis using magnetic resonance imaging (MRI). METHODS: This retrospective study was carried out in the radiology department of a tertiary healthcare center and 1.5-T MRI views of the shoulder obtained from 203 patients aged 8 to 29 years were evaluated. The ossification stage of proximal humeral epiphysis was evaluated on T1-weighted MRI sequences, and its correlation with age was sought. RESULTS: Our series consisted of 77 women (37.9%) and 126 men (62.1%). One hundred forty-seven right humerus (72.4%) and 56 (27.6%) left humerus were examined. The mean age of the patients was 20.60 ± 4.41 years. The patients were in stages 4 (n = 148, 72.9%), 3c (n = 18, 8.9%), 3b (n = 14, 6.9%), 2c (n = 13, 6.4%), and 3a (n = 10, 4.9%). Analysis of the relationship between age and ossification stage of proximal humerus epiphysis indicated that stage increased with the advancement of age. There was a significant correlation between the stage and age variables at a rate of 77% in the positive direction (P < 0.001). CONCLUSIONS: Our results demonstrated that MRI of proximal humeral epiphysis can have valuable implications for estimation of the skeletal age. Future prospective studies must be implemented with patients grouped according to socioeconomic status, nutritional habits, and physical activities to explore the actual investigative potential of MRI.

Estimation of sex using morphometric measurements performed on cranial computerized tomography scans.

We aimed to evaluate the usefulness of morphometric measurements performed on cranial computerized tomography (CT) images for the estimation of sex. A retrospective study was performed in the radiology department of a tertiary care center using data collected from cranial CT scans of 616 Caucasian cases (307 women, 309 men) with an average age of 44.70 ± 16.43. The parameters under investigation consisted of maximum cranial length (MCL), minimum frontal breadth, bi-zygomatic breadth (BZB), parietal chord, maximum cranial breadth, bi-mastoid diameter (BIM) and the length of cranial base. Any statistically significant difference in terms of these parameters was found between males and females. In our series, women were remarkably older than men (47.56 ± 15.87 vs. 41.39 ± 16.43; p < 0.001). We observed that there was a statistically significant difference between males and females concerning all morphometric measurements and males displayed higher values in terms of all parameters (p < 0.001, for all). The variables with the most successful performance for discrimination of gender were BZB (89.2%), MCL (87.4%) and BIM (84.8%). The concomitant use of these morphometric measurements seems to improve the accuracy of sex estimation. We suggest that morphometric measurements performed on cranial CT images can be useful for the estimation of sex.

Computed tomography characteristics of multiple myeloma and other osteolytic metastatic bone lesions.

BACKGROUND: Differentiation of multiple myeloma (MM) from osteolytic metastatic (OM) bone lesions may be critical in patients with lytic bone lesions but can be challenging for radiologists. PURPOSE: To determine whether computed tomography (CT) can be used to distinguish between MM and other OM bone lesions. MATERIAL AND METHODS: In this retrospective study, 320 lesions of 207 patients diagnosed with MM or OM, based on biopsy or clinical examination, were evaluated. Eight qualitative features were evaluated by two radiologists blinded to the diagnoses. The chi-square and Fisher exact tests, and logistic regression analysis, were used to evaluate the relationships between the CT findings and diagnoses. RESULTS: High-density areas were more common in OM than MM lesions (85.2% and 19%, P < 0.001), as were perilesional sclerosis (38.9% vs. 13.2%, P < 0.001), heterogeneity (on non-contrast CT images, 60% vs. 19.1%, P < 0.001; on contrast enhanced CT images, 80.6% vs. 28.2%, P < 0.001), and ill-defined margins (34.6% vs. 9.1%, P < 0.001). Similarly, OM lesions showed high-density areas more than MM in evaluation of skeletal system subgroups (vertebrae, 93.8% vs. 29.8%, P < 0.0001; thoracic cage bones, 69.6% vs. 19.2%, P < 0.001; pelvic bones and sacrum, 84.8% vs. 7.7%, P < 0.001; peripheral skeletal bones, 81.5% vs. 8.3%, P < 0.001). Logistic regression analysis revealed that the presence of a high-density area in the lesion increased the probability of a metastasis 25.88-fold (R2 = 0.516, P < 0.001). CONCLUSION: MM and OM lesions can be differentiated by CT; OM lesions exhibit high- density areas.